Solutions Manual For Data
Visualization Exploring and
Explaining with Data 2nd
Edition By Jeffrey Camm,
James Cochran, Michael Fry,
Jeffrey Ohlmann (All Chapters
1-9, 100% Original Verified,
A+ Grade)
All Chapters Arranged
Reverse: 9-1
This is The Original Solutions
Manual For 2nd Edition, All
other Files in The Market are
Fake/Old/Wrong Edition.
, Solution and Answer Guide: Camm/Cochran/Fry/Ohlmann, Data Visualization - Exploring and Explaining with Data, 2nd Edition, © 2025,
9780357929766; Chapter 9: Telling the Truth with Data Visualization
Solution and Answer Guide
CAMM/COCHRAN/FRY /OHLMANN, DATA VISUALIZATION - EXPLORING AND EXPLAINING WITH DATA,
2ND EDITION , © 2025, 9780357929766; C HAPTER 9: TELLING THE TRUTH WITH DATA
VISUALIZATION
TABLE OF CONTENTS
End of Section Problem Solutions...............................................................................1
CONCEPTUAL.......................................................................................................................................1
APPLICATIONS.....................................................................................................................................9
END OF SECTION PROBLEM SOLUTIONS
CONCEPTUAL
1. Identifying Missing Data Which of the following Excel methods can be used to help identify missing
data? LO 1
i. Conditional formatting to highlight missing data values.
ii. Using the COUNTIF() function to find the number of missing data values.
iii. Sorting the data to find missing data values.
iv. All of the above.
Solution:
iv.
2. Data Errors. Which of the following is true regarding data errors? LO 1
i. A data error is always identified by a unique numerical value such as 9999999.
ii. Any data that are determined to be outliers should be considered data errors and should be removed.
iii. Identifying outliers in a data set can be helpful in uncovering data errors.
iv. Data errors occur only when data are collected manually.
Solution:
iii.
© 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly
accessible website, in whole or in part.
1
, Solution and Answer Guide: Camm/Cochran/Fry/Ohlmann, Data Visualization - Exploring and Explaining with Data, 2nd Edition, © 2025,
9780357929766; Chapter 9: Telling the Truth with Data Visualization
3. Types of Data Bias. Match each of the following type of data bias with the correct description. LO 2, 3
Type of Data Bias Description
Selection bias Occurs when subsets of data show a specific trend, but that trend
disappears, or reverses, when the data are aggregated.
Simpson’s paradox Occurs when a data set is made up of a disproportionately high number of
observations corresponding to positive outcomes from a particular event.
Survivor bias Occurs when data are drawn from a sample that had not been properly
randomized to represent the intended population.
Solution:
Type of Data Bias Description
Selection bias Occurs when subsets of data show a specific trend,
but that trend disappears, or reverses, when the
data are aggregated.
Simpson’s paradox Occurs when a data set is made up of a
disproportionately high number of positive
outcomes from a particular event.
Survivor bias Occurs when data are drawn from a sample that
had not been properly randomized to represent the
intended population.
4. Potential Bias in Weight-Loss Study. A clinical trial has been conducted to evaluate the efficacy of a
new drug to enable weight loss for obese patients. A pool of 249 obese individuals are chosen for the study.
Each study participant must track their weight at home daily to compute their body-mass index (BMI) and have
a clinical evaluation once per week at a local hospital over six months to complete the clinical trial. At the end
of six months, it is found that 47% of those who received the new drug completed the clinical trial. Those who
completed the clinical trial are found to have reduced their BMI by 3.2 kg/m2, on average, over six months.
Explain how these results could be affected by bias and how that could affect the data. LO 2
Solution:
This is an example of potential survivor bias. The results of the clinical trial only apply to those that completed
the six-month trial. We do not know the results of those that dropped out of the trial prior to six months. It is
possible that those who were losing more weight were more likely to complete the trial, so this could bias the
data to show greater weight-loss by those taking the new drug than is accurate.
5. Simpson’s Paradox in Baseball. In the sport of baseball, batting average is calculated by dividing the
number of hits a player achieves by the number of official at bats. The following table shows the batting
performance for Mike Legg and Edison Vasquez over two consecutive seasons. LO 3
Season 1 Season 2
Player Hits At Bats Hits At Bats
Mike Legg 14 56 192 589
Edison Vasquez 112 418 57 149
a. Calculate the batting average for Mike Legg and Edison Vasquez in Season 1. Which player has a higher
batting average in Season 1?
b. Calculate the batting average for Mike Legg and Edison Vasquez in Season 2. Which player has a higher
batting average in Season 2?
c. Calculate the batting average for Mike Legg and Edison Vasquez over the combined Seasons 1 and 2.
Which player has the higher batting average over Seasons 1 and 2?
d. Explain how the results in parts a through c illustrate Simpson’s paradox.
Solution:
© 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly
accessible website, in whole or in part.
2
, Solution and Answer Guide: Camm/Cochran/Fry/Ohlmann, Data Visualization - Exploring and Explaining with Data, 2nd Edition, © 2025,
9780357929766; Chapter 9: Telling the Truth with Data Visualization
a. Mike Legg batting average in Season 1 = = 0.250
Edison Vasquez batting average in Season 1 = = 0.268
So, Edison Vasquez has the higher batting average in Season 1.
b. Mike Legg batting average in Season 2 = = 0.326
Edison Vasquez batting average in Season 2 = = 0.382
So, Edison Vasquez has the higher batting average in Season 2.
c. Mike Legg batting average over Seasons 1 and 2 = (14 + 192) / (56+589) = 0.319
Edison Vasquez batting average over Seasons 1 and 2 = (112 + 57) / (418 + 149) = 0.298
So, Mike Legg has the higher batting average over Seasons 1 and 2.
d. The results in parts a through c illustrate Simpson’s paradox because the subsets of data by individual
season show one result (Edison Vasquez has the higher batting average in each of Season 1 and Season 2),
but when the data are aggregated over the two seasons, the result is reversed (Mike Legg has the higher
batting average over Seasons 1 and 2 combined).
6. Average Nominal Hourly Earnings. An economist is examining wage growth in the United States. She
has collected data on average nominal hourly earnings for workers in the United States between 2006 and 2020,
which are shown in the following chart. Her conclusion is that hourly earnings for workers in the United States
have been steadily growing since 2006. How might this conclusion be incorrect, and how can she modify these
data to investigate the appropriateness of this conclusion? LO 4
© 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly
accessible website, in whole or in part.
3
Visualization Exploring and
Explaining with Data 2nd
Edition By Jeffrey Camm,
James Cochran, Michael Fry,
Jeffrey Ohlmann (All Chapters
1-9, 100% Original Verified,
A+ Grade)
All Chapters Arranged
Reverse: 9-1
This is The Original Solutions
Manual For 2nd Edition, All
other Files in The Market are
Fake/Old/Wrong Edition.
, Solution and Answer Guide: Camm/Cochran/Fry/Ohlmann, Data Visualization - Exploring and Explaining with Data, 2nd Edition, © 2025,
9780357929766; Chapter 9: Telling the Truth with Data Visualization
Solution and Answer Guide
CAMM/COCHRAN/FRY /OHLMANN, DATA VISUALIZATION - EXPLORING AND EXPLAINING WITH DATA,
2ND EDITION , © 2025, 9780357929766; C HAPTER 9: TELLING THE TRUTH WITH DATA
VISUALIZATION
TABLE OF CONTENTS
End of Section Problem Solutions...............................................................................1
CONCEPTUAL.......................................................................................................................................1
APPLICATIONS.....................................................................................................................................9
END OF SECTION PROBLEM SOLUTIONS
CONCEPTUAL
1. Identifying Missing Data Which of the following Excel methods can be used to help identify missing
data? LO 1
i. Conditional formatting to highlight missing data values.
ii. Using the COUNTIF() function to find the number of missing data values.
iii. Sorting the data to find missing data values.
iv. All of the above.
Solution:
iv.
2. Data Errors. Which of the following is true regarding data errors? LO 1
i. A data error is always identified by a unique numerical value such as 9999999.
ii. Any data that are determined to be outliers should be considered data errors and should be removed.
iii. Identifying outliers in a data set can be helpful in uncovering data errors.
iv. Data errors occur only when data are collected manually.
Solution:
iii.
© 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly
accessible website, in whole or in part.
1
, Solution and Answer Guide: Camm/Cochran/Fry/Ohlmann, Data Visualization - Exploring and Explaining with Data, 2nd Edition, © 2025,
9780357929766; Chapter 9: Telling the Truth with Data Visualization
3. Types of Data Bias. Match each of the following type of data bias with the correct description. LO 2, 3
Type of Data Bias Description
Selection bias Occurs when subsets of data show a specific trend, but that trend
disappears, or reverses, when the data are aggregated.
Simpson’s paradox Occurs when a data set is made up of a disproportionately high number of
observations corresponding to positive outcomes from a particular event.
Survivor bias Occurs when data are drawn from a sample that had not been properly
randomized to represent the intended population.
Solution:
Type of Data Bias Description
Selection bias Occurs when subsets of data show a specific trend,
but that trend disappears, or reverses, when the
data are aggregated.
Simpson’s paradox Occurs when a data set is made up of a
disproportionately high number of positive
outcomes from a particular event.
Survivor bias Occurs when data are drawn from a sample that
had not been properly randomized to represent the
intended population.
4. Potential Bias in Weight-Loss Study. A clinical trial has been conducted to evaluate the efficacy of a
new drug to enable weight loss for obese patients. A pool of 249 obese individuals are chosen for the study.
Each study participant must track their weight at home daily to compute their body-mass index (BMI) and have
a clinical evaluation once per week at a local hospital over six months to complete the clinical trial. At the end
of six months, it is found that 47% of those who received the new drug completed the clinical trial. Those who
completed the clinical trial are found to have reduced their BMI by 3.2 kg/m2, on average, over six months.
Explain how these results could be affected by bias and how that could affect the data. LO 2
Solution:
This is an example of potential survivor bias. The results of the clinical trial only apply to those that completed
the six-month trial. We do not know the results of those that dropped out of the trial prior to six months. It is
possible that those who were losing more weight were more likely to complete the trial, so this could bias the
data to show greater weight-loss by those taking the new drug than is accurate.
5. Simpson’s Paradox in Baseball. In the sport of baseball, batting average is calculated by dividing the
number of hits a player achieves by the number of official at bats. The following table shows the batting
performance for Mike Legg and Edison Vasquez over two consecutive seasons. LO 3
Season 1 Season 2
Player Hits At Bats Hits At Bats
Mike Legg 14 56 192 589
Edison Vasquez 112 418 57 149
a. Calculate the batting average for Mike Legg and Edison Vasquez in Season 1. Which player has a higher
batting average in Season 1?
b. Calculate the batting average for Mike Legg and Edison Vasquez in Season 2. Which player has a higher
batting average in Season 2?
c. Calculate the batting average for Mike Legg and Edison Vasquez over the combined Seasons 1 and 2.
Which player has the higher batting average over Seasons 1 and 2?
d. Explain how the results in parts a through c illustrate Simpson’s paradox.
Solution:
© 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly
accessible website, in whole or in part.
2
, Solution and Answer Guide: Camm/Cochran/Fry/Ohlmann, Data Visualization - Exploring and Explaining with Data, 2nd Edition, © 2025,
9780357929766; Chapter 9: Telling the Truth with Data Visualization
a. Mike Legg batting average in Season 1 = = 0.250
Edison Vasquez batting average in Season 1 = = 0.268
So, Edison Vasquez has the higher batting average in Season 1.
b. Mike Legg batting average in Season 2 = = 0.326
Edison Vasquez batting average in Season 2 = = 0.382
So, Edison Vasquez has the higher batting average in Season 2.
c. Mike Legg batting average over Seasons 1 and 2 = (14 + 192) / (56+589) = 0.319
Edison Vasquez batting average over Seasons 1 and 2 = (112 + 57) / (418 + 149) = 0.298
So, Mike Legg has the higher batting average over Seasons 1 and 2.
d. The results in parts a through c illustrate Simpson’s paradox because the subsets of data by individual
season show one result (Edison Vasquez has the higher batting average in each of Season 1 and Season 2),
but when the data are aggregated over the two seasons, the result is reversed (Mike Legg has the higher
batting average over Seasons 1 and 2 combined).
6. Average Nominal Hourly Earnings. An economist is examining wage growth in the United States. She
has collected data on average nominal hourly earnings for workers in the United States between 2006 and 2020,
which are shown in the following chart. Her conclusion is that hourly earnings for workers in the United States
have been steadily growing since 2006. How might this conclusion be incorrect, and how can she modify these
data to investigate the appropriateness of this conclusion? LO 4
© 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly
accessible website, in whole or in part.
3
